diff --git a/docs/docs/usage-guide/changing_a_model.md b/docs/docs/usage-guide/changing_a_model.md
index 420f3736a..d221b953c 100644
--- a/docs/docs/usage-guide/changing_a_model.md
+++ b/docs/docs/usage-guide/changing_a_model.md
@@ -201,3 +201,6 @@ fallback_models=["custom_model_name"]
 custom_model_max_tokens= ...
 ```
 (3) Go to [litellm documentation](https://litellm.vercel.app/docs/proxy/quick_start#supported-llms), find the model you want to use, and set the relevant environment variables.
+
+(4) Most reasoning models do not support chat-style inputs (`system` and `user` messages) or temperature settings. 
+To bypass chat templates and temperature controls, set `config.custom_reasoning_model = true` in your configuration file.
diff --git a/pr_agent/algo/ai_handlers/litellm_ai_handler.py b/pr_agent/algo/ai_handlers/litellm_ai_handler.py
index 0ef29b51d..b22b834e0 100644
--- a/pr_agent/algo/ai_handlers/litellm_ai_handler.py
+++ b/pr_agent/algo/ai_handlers/litellm_ai_handler.py
@@ -205,7 +205,7 @@ async def chat_completion(self, model: str, system: str, user: str, temperature:
                                           {"type": "image_url", "image_url": {"url": img_path}}]
 
             # Currently, some models do not support a separate system and user prompts
-            if model in self.user_message_only_models:
+            if model in self.user_message_only_models or get_settings().config.custom_reasoning_model:
                 user = f"{system}\n\n\n{user}"
                 system = ""
                 get_logger().info(f"Using model {model}, combining system and user prompts")
@@ -227,7 +227,7 @@ async def chat_completion(self, model: str, system: str, user: str, temperature:
                 }
 
             # Add temperature only if model supports it
-            if model not in self.no_support_temperature_models:
+            if model not in self.no_support_temperature_models and not get_settings().config.custom_reasoning_model:
                 kwargs["temperature"] = temperature
 
             if get_settings().litellm.get("enable_callbacks", False):
diff --git a/pr_agent/settings/configuration.toml b/pr_agent/settings/configuration.toml
index 01c09207e..ad09243a1 100644
--- a/pr_agent/settings/configuration.toml
+++ b/pr_agent/settings/configuration.toml
@@ -17,6 +17,7 @@ use_global_settings_file=true
 disable_auto_feedback = false
 ai_timeout=120 # 2minutes
 skip_keys = []
+custom_reasoning_model = false # when true, disables system messages and temperature controls for models that don't support chat-style inputs
 # token limits
 max_description_tokens = 500
 max_commits_tokens = 500